# Captivate: The Science of Succeeding with People — Work the Book
## Reading routes and assessment
Captivate organizes fourteen social-skills lessons around the first five minutes, five hours and five days of a relationship. **System assessment:** its conversation prompts, attention to other people's preferences and permission to decline invitations offer usable practice material. Its promises of universal decoding exceed the evidence evaluated here. Chapters 3,6,9 and 13 supply the main test: you can practice a question without claiming to know a stranger's dopamine level, hidden truth or single motive.
Read Prompt 7 for the verdicts, Prompt 0 for source tracing, and Prompts 11–12 for optional practice. This note completes the analytical chain with selective coverage; it doesn't claim a full reading of all fourteen chapters.
## Session record and book overview
Vanessa Van Edwards presents herself in the introduction as an awkward person who turns research into experiments and social-skills teaching through Science of People. That is the book's account of her work, not an independent credential audit. Her method combines cited psychology, interviews, personal stories, opt-in polls and her team's experiments. The argument asks you to treat interpersonal skill as something you can learn through practice.
The [publisher's page](https://www.penguinrandomhouse.com/books/539052/captivate-by-vanessa-van-edwards/) confirms the original hardcover and ebook publication on April 25, 2017 and a later paperback on June 19, 2018. This EPUB gives copyright 2017, ebookISBN9780399564505 and hardcover ISBN 9780399564482. Its March 29 OPF timestamp describes the package; it isn't substituted for the publication date. The [Amazon hardcover destination](https://www.amazon.com/dp/0399564489) identifies the matching title, author and hardcover ISBN. No purchase or Kindle change occurs.
**One-note adaptation:** source excerpts, analysis, literature notes, permanent-note candidates, teaching material, experiments and Mermaid remain in this note. The explicit user request overrides the older two-file rule and the spec's separate-note defaults. No separate drawing, Reader publication, permanent note or MOC edit occurs. Rating and personal reading dates remain unknown. Session dates record this processing, not Courtney's reading history.
# Prompt 0 — Retrieval and vault setup
## Identity, acquisition and integrity
The required mounted-file check returns “No such file or directory.” The exact-ID connector response matches the queue basename and gives 18,930,832 bytes. A temporary download completes in 1.12 seconds. SHA-256 is `d2003e5fed6aa13334b35034822ed8d9b61f9a1be62eac1d2d34693cc0d2ff97`. The ZIP CRC check passes all 168 entries; all 49 spine documents parse as XML. These checks test the package, not the author's conclusions. The canonical archive search returns no existing Captivate/Van Edwards review before creation. No source edit, move or rename occurs.
The on-disk book-note rule has the `hidden` flag and no `dataless` flag, so this run reads it. A Reader title-and-author search returns two unrelated article/email records and no verified Captivate book. This bounded search doesn't prove global absence. No matching reader highlight, color, timestamp, note or Readwise ID is available to deduplicate. **Layer 1 substitution:** the EPUB supplies the source. **Reader emphasis is unavailable.** All sixteen selections below reflect system judgment, not Courtney's endorsement.
## Coverage and source index
Read in full: extracted text in the introduction; chapters 3(Spark),6(Decode),9(Value),13(Protect); the three part introductions; copyright; What's Next?; and Notes. These eleven spine units contain approximately27,669 leaf-block words. Sampled: the opening paragraphs and chapter-review sections of chapters 1,2,4,5,7,8,10,11,12 and 14, approximately2,769 words. These samples establish the advertised project, not the full chapter's supporting evidence. Unread: the remaining paragraphs in those ten chapters, dedication, contents, digital bonuses, corporate workshops, flash-card appendix text, acknowledgments, index, separate footnotes and promotional final unit. Their claims and limitations are **not evaluated**.
All 49 spine units are mechanically extracted, totaling approximately 80,388 leaf-block words. The counts include repeated navigation and back matter and aren't a publisher word count. Extraction doesn't equal reading. Two image pixels are viewed: `OEBPS/image/37_Bogota_graffiti.jpg` and `OEBPS/image/56_Anger_Graphic.jpg`. Every other image, including the image-only cover/title pages and the microexpression appendix, remains not evaluated. The photographed anger diagram shows labeled facial movements; it doesn't validate an emotion or lie detector.
IDs use zero-based spine unit and body-element indices, for example `CAP-U013-e171`, plus the original package path. They aren't Kindle locations. In the analysis, `U013e171` abbreviates the same registered `CAP-U013-e171` unit and element. Quotes retain exact wording and punctuation; whitespace normalization is the sole comparison adjustment. Ellipsis-free fragments below quote contiguous source text. All sixteen pass native-text comparison.
| ID | Exact source excerpt | Location | Attribution |
| --- | --- | --- | --- |
| S-01 | “refined by thousands of students” | `CAP-U005-e021` · `OEBPS/xhtml/06_Introduction.xhtml` | Van Edwards; system-selected exact fragment |
| S-02 | “universal framework” | `CAP-U005-e035` · `OEBPS/xhtml/06_Introduction.xhtml` | Van Edwards; system-selected exact fragment |
| S-03 | “my goal is to help you level up your score by at least fifty points.” | `CAP-U005-e076` · `OEBPS/xhtml/06_Introduction.xhtml` | Van Edwards; system-selected exact fragment |
| S-04 | “People with high PQ make on average $29,000 more per year than people with average PQ.” | `CAP-U005-e058` · `OEBPS/xhtml/06_Introduction.xhtml` | Van Edwards; system-selected exact fragment |
| S-05 | “Use unique talking points to create pleasurable and memorable conversations.” | `CAP-U009-e042` · `OEBPS/xhtml/10_Chapter_3_Spark_How_t.xhtml` | Van Edwards; system-selected exact fragment |
| S-06 | “we randomly partnered over three hundred participants to test conversation starters.” | `CAP-U009-e087` · `OEBPS/xhtml/10_Chapter_3_Spark_How_t.xhtml` | Van Edwards; system-selected exact fragment |
| S-07 | “When any of these engagement cues happen, you know that you have just sparked some dopamine and pushed a hot button.” | `CAP-U009-e190` · `OEBPS/xhtml/10_Chapter_3_Spark_How_t.xhtml` | Van Edwards; system-selected excerpt |
| S-08 | “253 got deals and 242 did not.” | `CAP-U009-e269` · `OEBPS/xhtml/10_Chapter_3_Spark_How_t.xhtml` | Van Edwards; system-selected exact fragment |
| S-09 | “Spot the seven microexpressions during interactions to uncover truth.” | `CAP-U013-e171` · `OEBPS/xhtml/14_Chapter_6_Decode_How_.xhtml` | Van Edwards; system-selected exact fragment |
| S-10 | “Microexpressions (less than a second) cannot be controlled, so they are honest views into emotions.” | `CAP-U013-e181` · `OEBPS/xhtml/14_Chapter_6_Decode_How_.xhtml` | Van Edwards; system-selected excerpt |
| S-11 | “People can have disdain for an idea or a situation while still having a great relationship with you.” | `CAP-U013-e342` · `OEBPS/xhtml/14_Chapter_6_Decode_How_.xhtml` | Van Edwards; system-selected excerpt |
| S-12 | “If you notice someone using one of the seven microexpressions frequently along with their words, it might just be their facial punctuator.” | `CAP-U013-e468` · `OEBPS/xhtml/14_Chapter_6_Decode_How_.xhtml` | Van Edwards; system-selected excerpt |
| S-13 | “The underlying motivation that drives a person’s decisions, actions, and desires.” | `CAP-U016-e140` · `OEBPS/xhtml/17_Chapter_9_Value_How_t.xhtml` | Van Edwards; system-selected exact fragment |
| S-14 | “Some of my students have found the need to occasionally add a situational value.” | `CAP-U016-e502` · `OEBPS/xhtml/17_Chapter_9_Value_How_t.xhtml` | Van Edwards; system-selected exact fragment |
| S-15 | “Sometimes, as with many problems, they cannot fully be “fixed.”” | `CAP-U021-e445` · `OEBPS/xhtml/22_Chapter_13_Protect_Ho.xhtml` | Van Edwards; system-selected exact fragment |
| S-16 | “I won’t be able to get together.” | `CAP-U021-e545` · `OEBPS/xhtml/22_Chapter_13_Protect_Ho.xhtml` | Van Edwards; system-selected exact fragment |
## Inferred cluster map
| Cluster | Read material | Topic and rationale | Representative selections |
| --- | --- | --- | --- |
| A | Introduction U005 | Method, scope and promised outcomes; tests the contract behind later advice | S-01–S-04 |
| B | Chapter 3 U009 | Conversation, observable ratings and inferred brain mechanism; separates practical results from explanation | S-05–S-08 |
| C | Chapter 6 U013 | Facial interpretation, certainty and native exceptions; tests the decoding promise | S-09–S-12 |
| D | Chapters 9/13 U016/U021 | Motives, situational variation, support and boundaries; tests whether labeling respects agency | S-13–S-16 |
The clusters are system-inferred, not the author's parts. Selection density is 4 excerpts per cluster and measures this audit design. Repeated themes include prediction, novelty, appreciation, primary value, fear and connection. No reader-interest ranking follows from that density.
# Prompt 1 — Setup, contract, baselines, biases and priors
> [!note] System-inferred reading contract
> Examine which practices you can try as respectful conversation and which explanations require more evidence. Courtney hasn't supplied personal motives, prior beliefs, social difficulties or intended applications. Correct this contract before treating it as yours.
Success means separating observed behavior from inferred motive, tracking evidence dependence and preserving counterexamples. A controlled transfer study with defined outcomes and representative participants could change the judgments below. The selected source can't establish universal success, diagnosis, concealed truth or a causal income gain. Constraints include selective chapter reading, unviewed figures, unavailable raw laboratory datasets and absent reader emphasis.
**System baseline, not reader biography:** ordinary questions can help a conversation; no one cue warrants a complete personality judgment. The analyst's incentive to finish the queue can favor neat summaries; Prompts 6–7 counter this by keeping weak claims and native exceptions in view. The author's book, courses and workshops share an incentive to make practice sound predictable. That incentive doesn't prove dishonesty. No personal incentive or belief is assigned to Courtney.
## Open questions
| ID | Question |
| --- | --- |
| O01 | What does “universal” mean: all people, common patterns or adaptable suggestions? |
| O02 | Does PQ measure book-answer recall or interpersonal performance? |
| O03 | Which observations count as a spark without guessing dopamine? |
| O04 | What distinguishes interest from polite participation? |
| O05 | Does speed networking counterbalance question order? |
| O06 | Do three-minute ratings predict future friendship? |
| O07 | What rate of unsuccessful pitches also uses novelty? |
| O08 | Can a facial movement identify an emotion without context? |
| O09 | Can uncontrollability establish truthfulness? |
| O10 | How does the method handle punctuators and sun-induced squinting? |
| O11 | Is a primary value stable across contexts and time? |
| O12 | Can resource categories overlap in the same interaction? |
| O13 | Who codes the 100 diaries and with what agreement? |
| O14 | What besides fear can explain difficult behavior? |
| O15 | When should listening stop and a boundary begin? |
| O16 | Who can refuse touch, gaze or a personal question without penalty? |
| O17 | How do power, culture, disability and access affect transfer? |
| O18 | What new evidence or platform changes alter the evaluation? |
## System priors, declared before verdicts
These are reconstructed analytical starting judgments, not preregistered predictions and not Courtney's beliefs. The ordered prompt chain makes them explicit without inventing a prior reading history.
| Claim | Initial belief | Confidence | Reason | Evidence needed |
| --- | --- | --- | --- | --- |
| C-001 | A universal social recipe is unlikely | Low | Context changes goals | Cross-context prospective tests |
| C-002 | Students can help refine material | Moderate | Feedback can reveal usability | Sampling, failures and revisions |
| C-004 | A quiz can teach book vocabulary | Moderate | Recall responds to teaching | Independent performance criterion |
| C-005 | Fresh questions can invite engagement | Moderate | They open more specific topics | Comparisons and opt-out outcomes |
| C-008 | Novel pitches may associate with deals | Uncertain | Product and editing can confound | Novelty rates in both outcomes |
| C-009 | Expressions sometimes convey emotion | Moderate | Faces communicate | Dynamic contextual validation |
| C-015 | One stable motive per person is doubtful | Low | Needs compete | Repeated measurements and alternative models |
| C-018 | Naming feelings can invite clarification | Moderate | It can convey attention | Consent, correction and comparative outcomes |
**Drift:** the 2017 words and historical samples won't change. Research evaluation, online platforms, available bonus resources and communication norms can change. No current salary projection or live dating-site reply rate follows from a 2017 illustration.
# Prompt 2 — Inspectional reading
## Native book map
| Native chapter | Role in the advertised argument | Coverage |
| --- | --- | --- |
| Introduction | Teaching promise, PQ quiz and method | Full text |
| PartI: The First Five Minutes | First contact | Part introduction full |
| 1 Control: How to win the social game | Choose settings and energy | Opening/review sampled |
| 2 Capture: How to make a killer first impression | Nonverbal first impression | Opening/review sampled |
| 3 Spark: How to have dazzling conversations | Questions, novelty, names | Full text |
| 4 Highlight: How to be the most memorable person in the room | Listening and expectations | Opening/review sampled |
| 5 Intrigue: How to be ridiculously likable | Shared interests | Opening/review sampled |
| PartII: The First Five Hours | Understanding people | Part introduction full |
| 6 Decode: How to uncover hidden emotions | Facial reading | Full text |
| 7 Solve: How to crack someone's personality | Personality matrix | Opening/review sampled |
| 8 Appreciate: How to get the best from people | Appreciation preferences | Opening/review sampled |
| 9 Value: How to get along with anyone | Resource needs and motives | Full text |
| PartIII: The First Five Days | Continuing relationships | Part introduction full |
| 10 Connect: How to speak so people listen | Stories | Opening/review sampled |
| 11 Empower: How to lead people | Shared mission and ownership | Opening/review sampled |
| 12 Reveal: How to build lasting relationships | Vulnerability | Opening/review sampled |
| 13 Protect: How to deal with difficult people | Fear, listening and boundaries | Full text |
| 14 Engage: How to turn people on | Curiosity and belonging | Opening/review sampled |
| What's Next?; Notes | Practice follow-up and reference trail | Full text |
The native headlines name proposed benefits, not verified outcomes. Confidence is high for the sequence, limited for claims inside unread chapter bodies. The rhetorical modes are teach(practice challenges), persuade(universal promise), explain(brain mechanisms), narrate(interviews/stories) and sell(bonus/workshop links).
**Provisional summary:** Captivate presents interpersonal skill as learnable and divides progress into first impressions, deeper understanding and continuing relationships. Van Edwards moves between research summaries, biographies, personal experiments and exercises. In the chapters read here, she recommends specific questions, remembering names, observing responses, considering resource needs and listening before proposing solutions. She also gives stronger explanations: conversation produces dopamine, fleeting expressions reveal truth, and a primary value predicts choices. Her own text complicates these explanations through facial punctuators, accidental photo expressions, context-specific values and people who don't respond to empathy. The practical invitation to notice and ask survives those exceptions. The promise of a reliable person-decoding system needs evidence that matches ordinary, culturally varied interactions. This review treats the four fully read chapters as evidence and the ten opening/review samples as a map. It doesn't claim that unread sections lack qualifications or supporting studies.
**Hotspots:** introduction for generalization and test validity; chapter 3 for the link between ratings and neurotransmission; chapter 6 for truth claims; chapter 9 for stable motives; chapter 13 for empathy's limits; Notes for provenance. Deeper guiding questions are O01–O04, O05–O10, O11–O15 and O17: sixteen questions drawn from the inventory. **Recommended next reading, system inference:** selective deep reading of the ten remaining chapter bodies before using their tactics. Change to a full analytical read if the goal becomes assessing the entire training framework; stop a specific tactic when it requires imposing an unverified label on someone.
# Prompt 3 — Classification and terminology engine
This is popular self-development with business and relationship applications. The primary method is practical synthesis, not a systematic review. The following author meanings are **paraphrases** unless they point to the exact excerpt table. Competing and operational meanings are system proposals.
| Term | Author meaning | Competing meaning | Operational meaning for this audit | Risk | Example | Counterexample | Source |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Behavior hacking | Research-based shortcuts | Iterative skill learning | Named tactic with observable outcome | Formula suggests control | Try a question | Guaranteed success | U005e21–35 |
| PQ | Interpersonal intelligence | Ability separate from answer recall | Quiz score only | Score becomes trait |20×10 points | Income prediction from quiz | U005e75–260 |
| Universal | Framework for anyone | Common but bounded pattern | Explicit scope to test | Exceptions disappear | Varied settings | Single cultural template | S-02 |
| Spark | Pleasurable conversation peak | Moment of interest | Self-reported interest | Interest becomes dopamine | Topic invited | Polite nod | S-05/S-07 |
| Big Talk | Novel stimulating conversation | Mutual topic development | Participant-chosen topic | Small talk devalued | Shared hobby | Pressured disclosure | U009e29–35 |
| Hot button | Topic that lights someone up | Possible interest | Ask if they want more | Gesture treated as certainty | They request detail | Nervous laughter | U009e178–222 |
| Dopamine | Pleasure/memory explanation | Specific biological mechanism | Require biological measure for biological claim | Reverse inference | Brain study | Nod without assay | U009e44–81 |
| Microexpression | Brief involuntary facial response | Movement with uncertain context | Describe movement before inference | Emotion and truth conflated | Brow movement | Squint in sun | U013e40/181/502 |
| Congruency | Words match visible emotion | Multiple-channel consistency | Ask about discrepancy | Observer overrides speaker | Clarification | Accusation of lying | U013e176 |
| Facial punctuator | Movement accents speech | Habitual expressive style | Repeated pattern, no truth claim | Misclassification | Cowell example | Universal contempt attribution | S-12/U013e466 |
| Matrix | Personality/preferences/value profile | Provisional shared notes | Self-described preferences with correction | Snapshot freezes person | Ask preference | Hidden hiring dossier | U016e143–149 |
| Resource | Transferable relational good | Need, right or obligation | Name specific requested support | Rights become trade | Information requested | Wage replaced by praise | U016e29–116 |
| Primary value | Main driver | Several concurrent needs | Tentative priority in one context | One category explains all | Stated learning goal | Stability plus care | S-13 |
| Situational value | Different value in situation | Contextual priority | Record context and revisit | Exception rescues every prediction | Leadership recognition | Fixed category | S-14 |
| Gremlin | Social fear | One possible cause of distress | Fear named by speaker | Disability or injustice relabeled | Fear of rejection | Unpaid wage complaint | U021e21–64 |
| Name, Understand, Transform | Respond to emotion then help | Consent-based listening | Check feeling, listen, ask desired help | Amateur diagnosis | “Did I understand?” | Instruct someone to calm down | U021e180–520 |
**Contested language:** “truth,” “universal,” “primary” and “hacking” promise greater certainty than the exceptions support. “Toxic” and “passive” can function as condemnation; they aren't diagnoses. **Category-error watchlist:** correlation→causation present in pitch inference(C-008); anecdote→population present in stable values(C-015); behavior→identity present as risk in difficult-person types(C-017); intention→effect present when flattering reassurance stands in for repair in the safari story(U016e20); biological capability→practical outcome present in decoding(C-010); absence of evidence→evidence of absence is a risk this review avoids through coverage labels. Legal/moral conflation isn't established in the read sections; theology/history isn't applicable to the assessed claims.
# Prompt 4 — Claims inventory, dependencies and evidence expectations
EA=explicit author claim; Q3P=third-party claim reported by the author; RI=system inference. **RI rows aren't the author's sentences and must never appear as author quotations.** All claims are paraphrases. Locations refer to the native paths registered in Prompt 0.
| ID | Claim and source class | Type/scope | Test and expected evidence | Concrete falsifier | Terms; source; questions |
| --- | --- | --- | --- | --- | --- |
| C-001 | EA: fourteen hacks form a universal framework | Predictive; all social settings | Prospective tests across settings and access needs; not supplied in read introduction | Defined groups repeatedly fail despite correct use | Universal; U005e35–56; O01/O17 |
| C-002 | EA: thousands of students refine each skill | Descriptive; teaching program | Feedback sampling and revision trail; testimony supplied | Records show tactics unchanged despite failures or no feedback base | Method; S-01; O05 |
| C-003 | EA: high PQ earn$29,000 more per year on average | Descriptive; unspecified population | Defined PQ measure, pay data and adjusted comparisons; no matched dataset here | Matching population fails to reproduce difference | PQ; S-04; O02 |
| C-004 | EA: repeating PQ test shows improved interpersonal skill | Methodological; readers | External performance validation; scoring scheme supplied | Recall score rises while blind interaction ratings don't | PQ; U005e75–76/U023e8; O02 |
| C-005 | EA: unique talking points create enjoyable memorable conversations | Practical; conversations | Partner ratings and later recall; examples and ratings supplied | Partners report worse comfort/recall in matched trials | Spark; S-05; O03/O04 |
| C-006 | EA: engagement cues establish a dopamine spark | Causal; observed cues | Biological measures tied to interactions; absent in read report | Same cues occur without proposed signal or from other motives | Dopamine; S-07; O03/O04 |
| C-007 | EA: fresh prompts outrank conventional questions in networking | Descriptive; reported 2016 experiment | Counterbalanced ratings, uncertainty and raw data; ranks/sample description supplied | Controlled rerun reverses ranking | Big Talk; U009e87–107; O05/O06 |
| C-008 | EA: unique pitches have greater chance of a deal | Predictive; televised pitches | Novelty×deal contingency table and controls; success-side 63% supplied | Failure pitches use novelty equally/more often | Novelty; U009e269–274; O07 |
| C-009 | EA: seven universal expressions reveal hidden emotional truth | Causal/practical; all humans | Dynamic spontaneous contextual accuracy tests; classic studies narrated | Same movement reliably signals different states or non-emotion | Microexpression; S-09/U013e43; O08/O10 |
| C-010 | EA: subsecond expressions cannot be controlled and therefore are honest | Causal; fleeting facial movement | Timing/control tests and truth criteria; assertion supplied | Brief produced movement misleads or honest movement lacks diagnostic state | Truth; S-10; O09 |
| C-011 | EA: frequent apparent expressions may punctuate speech | Descriptive; particular speakers | Within-speaker dynamic observations; Cowell illustration supplied | Repeated pattern tracks internal state independent of speech in tested speaker | Punctuator; S-12; O10 |
| C-012 | EA: fake/closed-mouth smiles rate lower in lab's photo sample | Descriptive;400 coded photos | Coding agreement and model controlling image conditions; sample described | Matched analysis shows no difference | Rating; U013e494–503; O10 |
| C-013 | EA: primary value means underlying driver of choices | Definitional; proposed vocabulary | Conceptual usefulness; empirical validation separate | Definition itself not empirically falsifiable | Primary value; S-13; O11/O12 |
| C-014 | Q3P: Foa frames interactions as resource transactions | Interpretive; relationships | Original theory and competing interpretations; summary/citations supplied | Framework cannot distinguish exchange from obligation without ad hoc recoding | Resource; U016e29–36; O12 |
| C-015 | EA:100 diaries reveal a primary resource for each person | Descriptive/predictive; seven-day study | Coding reliability, validation and longer repeats; method narrative supplied | Independent coders fail agreement or held-out choices resist categories | Primary value; U016e155–240; O11/O13 |
| C-016 | EA: a person's value can differ by situation | Descriptive; context | Within-person repeated reports; examples supplied | Predefined contexts show stable priorities despite expected switches | Situational value; S-14; O11 |
| C-017 | EA: difficult behavior stems from fear | Causal; difficult interactions | Competing-cause analysis; anecdote and neuroscience summaries supplied | Deliberate harm, power incentives or material injustice explain behavior without fear | Gremlin; U021e21/173; O14 |
| C-018 | EA: naming, understanding and helping can calm and support | Practical; willing participants | Comparative interaction outcomes and boundary failures; scripts supplied | Correctly offered method repeatedly escalates willing partners | NUT; U021e180–522; O14/O15 |
| C-019 | RI: permission and correction strengthen a conversation practice | Normative; this review's adaptation | Explicit ethical principle; no effect estimate | N/A norm; failures can revise its proposed implementation | Agency; U021e356/517; O16/O17 |
| C-020 | RI: you can decline without accepting a personality diagnosis | Normative; this review's adaptation | Boundary principle; no prevalence claim | N/A norm; requires obligations/safety context | Boundary; S-16/U021e558; O15/O16 |
**Evidence pattern:** the fully read chapters provide examples, some sample descriptions, references and ranks. They don't supply the raw feedback, biological measures, comparative pitch table or predictive validation needed for the broadest assessed claims. That is a scoped finding about those reports; unread chapter bodies are not evaluated.
**Dependency outline:** learning/practice(C-002,C-004) → conversation options(C-005,C-007) → adaptable connection. The measured-cue→known-brain-state bridge(C-006), face→truth bridge(C-009,C-010) and diary→stable person bridge(C-015) support the broad universal claim(C-001) and can fail independently. Their failure doesn't eliminate asking a question or respecting a boundary. C-019/C-020 are system safeguards, not author's premises. C-003 salary association is separate from those practical outcomes.
# Prompt 5 — Quantitative rigor setup
Use qualitative updates. The source gives no comparable likelihood model for these 20 claims, and percentages from different outcomes can't produce a posterior probability of the entire framework. Confidence labels describe judgment, not measured probability. Multiple examples, student reports and Twitter polls from the same author organization count as one evidence family; references can lead to independent studies, but the citation itself doesn't verify them.
**Numbers checked against native text:**20 quiz answers×10 points yields 0–200; a 50 point target can reflect learning answers. The pitch counts 253+242=495 and 253/495≈51.1%. The 63% figure is the fraction of successful pitches using uniqueness; it doesn't give success conditional on uniqueness. Without the unsuccessful-group rate, the predictive comparison is incomplete. “Over 300” participants, seven questions and three-minute rounds don't establish counterbalancing or independent observations. The 100 diarists(50 men/50 women) produce repeated entries over seven days, not 700 independent people. A binary recruitment split doesn't establish general gender representation.
| Warning | Applicability/severity | Reason |
| --- | --- | --- |
| Correlated sources | High | One lab, repeated students/diaries |
| Missing denominators | High | Income population and poll response totals unresolved |
| Selection effects | High | Opt-in digital surveys and TV selection |
| Reverse biological inference | High | Nods don't measure neurotransmission |
| Proxy outcome | High | Quiz recall, attractiveness and likes differ from relationship quality |
| Base-rate neglect | High | Successful pitch percentage lacks comparison group |
| Survivorship | Moderate | Memorable successes dominate anecdotes |
| Publication bias | Unresolved | Complete search and unpublished studies not reviewed |
| Tiny sample | Unresolved | Stated headcount alone doesn't reveal subgroup power |
| P-value treated as effect size | Not assessed | No such calculation evaluated here |
| Statistical vs practical effect | Moderate | Good immediate ratings don't establish durable benefit |
Prompt 7 applies increases/decreases only after assessing evidence. Definition and normative claims receive N/A, not fake numerical confidence.
# Prompt 6 — Analytical reading scaffold
## Filled cluster analyses
**A: contract and validation.** Purpose: make social practice feel learnable. Thesis: a tested framework changes interactions. Claims C-001–004; terms PQ/framework; evidence personal history/student feedback/quiz. Example: twenty items and a fifty-point goal. Counterexample: the introduction reports a chimp-posture experiment that doesn't work(U005e17). Assumption: course performance transfers. Unanswered: O02 and salary measurement. Spine role: validates the teaching promise. Sources S-01–04. **System response:** retain learnability while asking for external criterion validation.
**B: conversation and novelty.** Purpose: replace routine prompts with interest. Claims C-005–008; terms Spark/dopamine; evidence Bogotá guide, soup conversation, networking ratings and TV coding. Counterexample **not supplied as a tested failure** in the read novelty report; the review's privacy-sensitive question scenario is an inference. Assumptions: enjoyment, recall and neurochemistry align. Questions O03–07. Spine role: practical first contact. Sources S-05–08. **System response:** the question can stand without its biochemical explanation.
**C: facial decoding.** Purpose: notice responses and adjust support. Claims C-009–012; terms microexpression/punctuator; evidence Ekman summaries, televised dating,400 photos and worked scripts. Native counterexamples include squinting in sun, flash reactions, punctuators and non-personal contempt. Assumption: observer identifies the correct trigger. Questions O08–10. Spine role: strong prediction bridge. Sources S-09–12. **System response:** treat movements as prompts for clarification; edited TV can't verify a hidden cause.
**D: values and difficulty.** Purpose: consider needs and listen before solving. Claims C-013–018 plus RIs 019–020; evidence Foa summary,100 diaries, safari stories, emails and invented dialogues. Native counterexamples: situational values, empathy that doesn't work, problems that can't be fixed. Assumption: six categories and fear cover the relevant motives. Questions O11–17. Spine role: longer-term support. Sources S-13–16. **System response:** ask for a person's account and preserve refusal; the elephant anecdote's improved guest mood doesn't independently assess safety responsibility.
For the ten sampled chapters, Prompt 2 records purpose and sources; their full evidence, assumptions, counterexamples and conclusions remain **not evaluated**. A chapter-review slogan isn't a substitute for its whole argument. No reader reaction is fabricated.
## Argument spine and evidence ledger
System reconstruction: people can practice social skill → select contexts and conversational opportunities → attend to feedback → adapt support → deepen relationships. The uncertain step is treating interpreted feedback as a decoded hidden state. **Weakest link:** confident universal diagnosis from cues(C-001/C-009/C-010). Repair needs dynamic tests, independent criteria, error rates, alternatives and cultural/access boundaries; narrowing to a question sacrifices the decoding promise but preserves attention.
| Evidence | Claims | Source/type | Independence and quality | Limits |
| --- | --- | --- | --- | --- |
| E1 | All EA/Q3P | EPUB text,16 exact excerpts | One author synthesis; high fidelity to wording | No independent verification of reported participants/results |
| E2 | C-005/C-007; RI019 | [Huang et al.,2017](https://www.hbs.edu/ris/Publication%20Files/Huang%20et%20al%202017_6945bc5e-3b3e-4c0a-addd-254c9e603c60.pdf), abstract and selected discussion pp 1,12–13 | Independent research report, three studies | Questions/follow-ups and liking; not a test of this book's exact prompts or dopamine. Field results are correlational; asker controls matter. Full methods, data and current replication record not audited |
| E3 | C-009/C-010 | [Barrett et al.,2019](https://journals.sagepub.com/doi/abs/10.1177/1529100619832930), publisher abstract | Independent review team; challenges a fixed facial mapping | Abstract supports variation across contexts/cultures and multiple meanings. Full review not read; no proof all emotion recognition fails |
| E4 | C-009/C-011 | [Keltner et al.,2019](https://pmc.ncbi.nlm.nih.gov/articles/PMC6687086/), selected multimodal/recognition sections | Independent theoretical research synthesis | Describes above-chance recognition and broader multimodal expression. Supports information in behavior, not guaranteed hidden truth; full paper and original studies not audited |
Three external research records are independently authored; they aren't three independent replications of Captivate. E3/E4 draw on overlapping emotion literatures and competing interpretations. E2 narrows conversation support. E3 tells against the strong decoding inference; E4 prevents replacing that criticism with “faces convey nothing.” These bounded checks don't constitute a complete 2026 literature review. Publisher/Amazon checks establish identity, not efficacy.
**Reading plan and gate:** full source passages for every assessed claim, stable terms, falsifiers, source-role separation and exceptions are present before Prompt 7. Priority next reading is the unexamined personality/appreciation bodies and the original studies behind any tactic you intend to use. Compare definitions, stop after each method/outcome distinction, and allow sustained attention for neuroscience and measurement. No completion-time promise follows from this plan.
# Prompt 7 — X-ray analysis and fair criticism
**Strongest reconstruction, system inference:** Van Edwards wants you to stop treating social competence as an innate gift, choose settings that fit, become curious and adjust support to others. Her exercises convert abstract reading into attempts you can inspect. That project survives without a universal decoding machine. Evidence can support responsiveness while leaving the proposed neurotransmitter or motive explanation uncertain.
| Claim | Support and grade | Strongest objection/counterexample | Current conclusion and repair |
| --- | --- | --- | --- |
| C-001 | **F**; broad promise plus anecdotes | No all-settings validation; native exceptions | Universal efficacy unsupported here. Repair as adaptable practice menu; loses guarantee |
| C-002 | **C**; explicit feedback account | Missing participation/failure denominator | Refinement plausible, no independent quality proof. Publish feedback/revision trail |
| C-003 | **D**; isolated income figure | PQ population/measure unspecified | Association unverified; causal gain rejected as an inference. Supply source/population |
| C-004 | **D**; quiz and retest | Learning the answer key can raise score | Recall indicator, not validated social intelligence. Add blind real-interaction criterion |
| C-005 | **C**; examples and ratings; E2 adjacent | A fresh question can intrude | Try with permission; no universal enjoyment/recall claim. Narrow to willing contexts |
| C-006 | **D**; scientific narrative | Visible cue lacks biochemical specificity | No dopamine finding from conversation cue. Repair as reported interest, remove assay claim |
| C-007 | **C**; >300 participants and ranks | Order/repetition/ratings lack analysis | Scoped lab report, no universal best opener. Counterbalance and release distributions |
| C-008 | **D**;495 pitches,63% successful novelty | Missing unsuccessful novelty rate; product/editing confounds | Prediction unestablished. Publish two-outcome table and held-out validation |
| C-009 | **F**; classic narratives and illustrations | E3 and native punctuators challenge certainty; E4 supports bounded information | Hidden-truth decoder misleading as stated. Repair to context-sensitive hypothesis; loses certainty |
| C-010 | **F**; timing assertion | Involuntary doesn't logically imply diagnostic truth | Timing alone can't establish honesty. Needs controlled accuracy evidence; not repaired by renaming fast movements |
| C-011 | **C**; recurring-speaker example | No frequency/error study | Useful possible alternative explanation. Keep provisional and ask |
| C-012 | **D**;400 photo report | Attractiveness/site selection differs from relationship quality | Historical sample assertion, not universal photo rule. Publish coding/model and scope |
| C-013 | **N/A**; author's definition | Definition doesn't validate prediction | Vocabulary available; test subsequent claims separately |
| C-014 | **C**; Foa summary and examples | Need, duty and resource can overlap | Interpretive lens, not exhaustive cause. Original theory not independently read; retain Q3P |
| C-015 | **D**;100 seven-day diaries | Categories/coding and stable predictive accuracy unavailable | Single driver not established. Repair to participant-reported contextual priority |
| C-016 | **C**; situational examples | Exception could absorb any failed prediction | Text concedes variation. Predefine change conditions before claiming prediction |
| C-017 | **D**; wedding story and fear account | Injury, incentives or unfairness needn't reduce to fear | One possible explanation. Narrow scope and don't absolve harm |
| C-018 | **C**; scripts and acknowledged failures | Feeling misnamed or premature solution escalates | Optional listening scaffold. Ask/check/allow correction; no hostage/clinical protocol |
| C-019 | **N/A**; explicit system norm | Permission alone doesn't guarantee comfort | Ethical safeguard, not efficacy result or author quotation |
| C-020 | **N/A**; explicit system norm | Duties and immediate safety can change choices | Boundary principle with context; no diagnosis required |
**Distribution:** A 0, B 0, C 7, D 7, F 3, N/A 3 = 20. Seven C grades mean suggestive/incomplete support, not 7 confirmed outcomes. Seven D grades mark measurement or transfer gaps;3 F grades concern absolute decoding promises. Norms/definitions receive no empirical grade. The limited external reading and selective chapter coverage prevent a whole-book evidence verdict.
## Prior updates and reasoning audit
| Claim | Prior → post-evaluation judgment | Dominant evidence and dependence | Update |
| --- | --- | --- | --- |
| C-001 | Low → Low | Native exceptions and absent all-settings test | Decrease confidence in universal wording |
| C-002 | Moderate → Uncertain | One lab's self-report | Decrease; refinement not independent validation |
| C-004 | Moderate for recall → Moderate for recall, Low for transfer | Same quiz twice | No recall update; decrease transfer certainty |
| C-005 | Moderate → Moderate with conditions | E2 plus scoped networking report | Small increase for asking/following up; exact novelty remains C |
| C-008 | Uncertain → Low | Success-side percentage only | Decrease predictive confidence |
| C-009 | Moderate for emotion information → Moderate; Low for decoding | E3/E4 plus native exceptions | Retain information; decrease fixed truth mapping |
| C-015 | Low → Low | Single seven-day coded sample | No meaningful increase in stability |
| C-018 | Moderate → Moderate with permission | Scripts and explicit nonresponse | No population-effect estimate; preserve option |
These labels encode the analyst's judgment; no posterior probability is calculated. Each row states which scope changed.
**Warranted bias findings:** opt-in selection matters because Notes U028e19 describes website volunteers; diverse volunteers don't imply representative sampling. The pitch conditional-probability error(C-008) blocks the reported prediction. C-006 uses reverse inference from behavior to chemistry. C-017 overgeneralizes an explanatory cause; its neuroscience references don't test every difficult person's motive. EditedTV(U013e168) encourages hindsight interpretation of a romantic outcome. The six-resource diagram is an interpretive framework, not a finding that every action has one cause.
**Counter-observations in the author's favor:** the chimp experiment fails; chapter 1's sampled review values different social styles; chapter 6 distinguishes punctuators and photographic accidents; Notes acknowledges fear/surprise confusion(U028e186); chapter 9 admits situational values and different professional/social/romantic priorities; chapter 13 says some problems can't be fixed and some people don't respond to empathy. These admissions narrow criticism and support flexible practice. They don't resolve the remaining absolute statements.
**Attribution:** author says the cited EA claims; author reports Foa's Q3P theory; the source shows words and described samples; **we infer** the repairs and ethical safeguards. **We conclude** that questions and clarification are usable options while hidden-state guarantees remain unestablished. Actual user outcomes, full unexamined chapters and the complete replication record remain uncertain.
# Prompt 8 — Systems, context and intellectual lineage
**Publication context:** the 2017 text uses dating websites, Twitter polls, reality TV, digital bonuses and business networking. These supply both examples and selection mechanisms. Its fast social-progress frame assumes opportunities to attend events and act on individual tactics. Structural conditions such as pay, role authority and access can constrain outcomes even when a conversation improves. That last statement is a system interpretation of transfer limits, not a surveyed outcome from the book.
**Documented lineage:** Notes explicitly cites Carnegie(chapter 4), Ekman/Friesen(chapter 6), the five-factor personality literature(chapter 7), Chapman(chapter 8), Foa(chapter 9), LeDoux/Goleman and Goulston(chapter 13). These references document intellectual borrowing; they don't imply endorsement by those scholars or validate each adaptation. Original books/papers on that list aren't all read here.
**Later context:** E2's online-first date is April 27, 2017, two days after the publisher's original book date; this review treats it as an external comparison, not evidence of author influence. E3/E4 are 2019 emotion debates. Their relation is evaluative, not causal history. Current-platform tactics require fresh checking before use. No new salary inference, personality diagnosis or hiring rule follows from a historical example.
**Incentives and systems map:** research → author synthesis → practice/feedback → book/courses/workshops; the same organization often selects the evidence and sells the application. Readability and sales incentives can favor tidy mechanisms. That mechanism is a possible bias source, not proof that the reported experiments are fabricated. A repair adds independent feedback, failure recording and participant correction.
# Prompt 9 — Inclusion, culture, accessibility and reading demands
**Textual strengths:** chapter 1's sampled review says charisma has different forms(U007e308); chapter 9 allows contextual values; chapter 13 offers text/email refusal and time to consider invitations(U021e558). These are concrete options. The read material doesn't demonstrate procedural protection against a manager's misuse of decoding labels; such protection remains **not established in evaluated sections**, not absent from every page.
**System-inferred risks:** interpreting a quiet colleague as passive(U021e113), a facial response as truthful, or leaving at 5 p.m. as a money motive(U016e400) can override disability access, caregiving duties or the person's account. The gendered promotion dialogue(U021e195–512) pairs upset woman and rationalizing man; it is a worked example, not a demonstrated population difference. Luxury safaris and business-card experiments assume resources; online surveys exclude people without access or interest. The reported 50 men/50 women diary sample gives no assessment of other gender categories.
| Transfer setting | Native practice | Adaptation proposed by this review | Boundary |
| --- | --- | --- | --- |
| Different conversational norms | Fresh questions | Offer familiar topics and choice to pass | Don't demand private disclosure |
| Neurodivergent or disabled communication | Gaze/facial cues | Ask preferred channel, pace and clarification | No gaze-based competence score |
| Hearing/noisy environments | Social context choice | Quiet space, written exchange, captions | Quiet participation isn't disinterest |
| Unequal workplace power | Value/motivation | Ask explicit work needs and compensation | Praise doesn't replace pay or consent |
| Limited money/digital access | Bonus tools and gifts | Use a printed prompt and ordinary conversation | No purchases required |
**Actual image audit, bounded:** the Bogotá mural image shows a person looking upward with a patterned head covering, jewelry and surrounding painted motifs; the source calls the depicted woman indigenous, but the photograph doesn't identify a community or artist. No further identity is inferred from appearance. The anger diagram shows a photographed face with dotted red markers and black labels for brow, eyes and lips. Both native image elements have empty alt text. Text extraction omits the labels in pixels; chapter prose supplies some movement descriptions. Those observations don't establish demographic balance or access across all illustrations. The other figures remain not evaluated.
**Reading-level estimate, system inference:** conversational sentence structure, direct instructions and worked dialogues fit an adult general reader. Reference density is higher: dopamine, amygdala, thalamus, neocortex, named theories and unfamiliar TV examples require context. The five-hour/day headings describe a teaching organization, not verified progress deadlines. Plain-language definitions, text alternatives and a “movement versus interpretation” distinction reduce that burden. Supplementary perspectives to seek are first-person accounts from people whose preferred communication differs, culturally specific conversation practices and independent measurement reviews; none is invented as already consulted.
# Prompt 10 — Notes pipeline inside this note
**Fleeting notes:** request the pitch comparison denominator; separate question novelty from responsiveness; revisit the 100 diary coding protocol; ask whether a quiet person wants another channel. These are system questions, not Courtney's reactions.
## LN-1 — A question and its proposed mechanism are separate
Van Edwards gives prompts and networking ranks(U009e87–107), then interprets engagement cues as dopamine(S-07). The observable conversation result and biological explanation need different evidence. Links: [[#PN-1 — Practice without a hidden-state guarantee]], C-005–007.
## PN-1 — Practice without a hidden-state guarantee
**System permanent-note candidate:** you can offer a conversation prompt and ask about comfort without claiming a neurotransmitter result. **Confidence:** high for the logical separation; outcome efficacy remains C. **Scope/conditions:** willing ordinary conversation, accessible channel, easy refusal. **Counterexample/stop:** an unwanted question is refused; stop or change topic. Support and reciprocal link: [[#LN-1 — A question and its proposed mechanism are separate]]. Proposed future filename: `A conversation result needs its own measure.md`; no file is created.
## LN-2 — Facial exceptions limit decoding
Van Edwards describes punctuators and accidental photo movements(U013e466–503) beside her hidden-truth promise(S-09/S-10). These exceptions require context. Links: [[#PN-2 — Ask before interpreting a cue]], C-009–012.
## PN-2 — Ask before interpreting a cue
**System permanent-note candidate:** describe an observation, ask an open clarification and accept correction. **Confidence:** moderate as a respectful practice; no guaranteed accuracy. **Scope/conditions:** low-stakes conversation; never substitute for diagnosis, consent, safety assessment or hiring criteria. **Counterexample/stop:** clarification itself intrudes; leave the topic. Support and reciprocal link: [[#LN-2 — Facial exceptions limit decoding]]. Proposed filename: `A cue invites clarification.md`; no file is created.
## LN-3 — A preference can change with context
Chapter 9 allows different professional/social/romantic values and a situational value(S-14/U016e345–370). Chapter 13 acknowledges problems that can't be fixed(S-15). Links: [[#PN-3 — Keep a stated preference revisable]], C-015–018.
## PN-3 — Keep a stated preference revisable
**System permanent-note candidate:** record what someone says they need for the present situation and allow them to revise it. **Confidence:** moderate; it is a design rule, not predictive validation. **Scope/conditions:** voluntary mutual planning; record only necessary information with permission. **Counterexample/stop:** a label persists after correction; remove it. Support and reciprocal link: [[#LN-3 — A preference can change with context]]. Proposed filename: `A stated preference has a context.md`; no file is created.
**Structure and adjacency:** the existing [[1. Projects/Book Review Completion/index]] provides the completion-project context; its file exists. Its content isn't treated as evidence about communication. Prompt 0's clusters and Prompt 4's dependency outline supply the internal theme map. ReciprocalLN↔PN anchors stay here; external MOC reciprocity is deferred under the one-note constraint. No new MOC or duplicate file appears.
**Don't attribute to Van Edwards:** RI claims C-019/C-020, all PN candidates, ethical adaptations, external-study summaries, conditional repairs, grade assignments and the invented teaching scenario. Foa's reported theory remains Q3P; third-party voices aren't converted into Van Edwards's empirical results.
# Prompt 11 — Teaching pack
**Audience:** adult readers comparing practical social-skills advice with its evidence. **Objectives:** distinguish observation from inferred state; identify the missing pitch denominator; offer clarification that permits correction; choose a boundary without diagnosing someone. **Prerequisites:** percentages with denominators and the distinction between question, observation and explanation. Vocabulary: spark, cue, primary value, proxy, source dependence and consent.
## Complete worked example
**Invented teaching case, not a reported interaction:** Riley and Jordan plan a voluntary reading-group discussion. Jordan looks away when Riley proposes a face-reading exercise. Riley doesn't announce that Jordan fears criticism. Riley says, “Would you prefer a written example or a spoken discussion? Passing is fine.” Jordan says, “Written works; I'm trying to take notes.” Riley provides the text and asks whether a five-minute silent start works. Jordan can accept, decline or change the plan. No claimed biological state or lesson outcome follows.
The observation is looking away. The proposed fear is an unverified inference. The supplied account concerns note taking. The negotiated change supplies an accessible participation option. **Connection to the book:** attention and adaptation survive as practice; S-09's hidden-truth guarantee isn't taught as a settled fact. **Connection to the analysis:** PN-2/PN-3 are proposals whose effects remain open to feedback.
**Sequence,30 minutes if activated:**5 minutes distinguish source/observation/inference in the case;5 minutes reconstruct C-008's missing comparison;10 minutes optional pairs rewrite one clarification with a pass option;5 minutes discuss exceptions S-12/S-14;5 minutes write a private boundary sentence. Written or solo participation remains available throughout.
**Practice:** classify “the brow moves,” “the person is angry” and “the person is hiding truth” as observation, interpretation and stronger inference. Then add the evidence each would need. Compare a new prompt with a follow-up to the answer; neither is automatically better in every setting. Discuss: what changes when one person controls pay or assessment? What if the clarification is declined? How would a situational value disconfirm a fixed matrix?
**Misconception checks:** a quiz score doesn't validate real-world transfer;63% among successful pitches doesn't give the success rate for unique pitches; a biological explanation doesn't establish that a tactic works; facial-information criticism doesn't mean faces convey nothing. The correct assessment identifies the missing evidence, not a preferred author or school.
**Rubric:**0 absent,1 partial,2 clear for four criteria: source attribution, denominator awareness, observation/inference separation and permission/correction. Maximum 8 points measures this exercise, not PQ or interpersonal worth. Learners supply their own responses; no scores or understanding are invented. Reflection: which claim do you retain after narrowing its scope, and which decision still needs evidence?
**Facilitator cautions:** disclose the four-chapter full read and ten-chapter sampling; don't use face images as tests of hidden emotion; don't ask for trauma disclosure or label classmates; don't treat the hostage analogy as a safety protocol; avoid stereotypes based on gaze, tone or fluency. Offer breaks, text, captions for any future audio, quiet space and no-touch participation. Adapt scenarios to local conversational norms through participant input; don't assume a cultural norm from appearance. Further reading is E2–E4 with the stated coverage limits, followed by original research relevant to the chosen claim.
# Prompt 12 — Action, PARA, experiments and drift
**Take, if you choose:** use a question that permits a pass; remember the distinction between cue and interpretation; inspect denominators. **Consider:** ask which kind of support someone wants; revisit the remaining chapters before adopting their tactics. **Avoid:** covertly profile a coworker, equate quietness with deficiency, infer deceit from a face or replace a pay dispute with appreciation. **Most likely misuse:** turning this review's limited skepticism into a new universal rule that you can judge people from their visible behavior.
**PARA:** this canonical review belongs in Archives as completed processing. The embedded terms can serve future Resources; the optional exercises can support an Area of communication if Courtney chooses. No new Project, commitment, scheduled review or file follows automatically.
All experiments below are **inactive proposals**, not assignments or completed tests. They examine PN transfers, not medical or personality interventions. Other people hear what you're trying and can decline.
| Experiment | Claim/PN | Hypothesis | Action | Measure | Duration | Stop | Review |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Source-role check | C-006/PN-1 | Separating explanation helps evidence judgment | Solo: mark 5 sentences as observation, inference or mechanism | Count justified labels; note uncertainty |10 minutes once | No located evidence; leave unresolved | At exercise end if activated |
| Optional question practice | C-019/PN-2 | A pass option supports comfort | Tell a willing partner you're trying a clearer invitation; ask one question | Partner's optional comfort feedback; no hidden trait score | One ordinary conversation | Any refusal or discomfort | After conversation, only if wanted |
| Context revision | C-016/PN-3 | Explicit preferences prevent a fixed label | Ask a willing collaborator preferred discussion format for one meeting | Whether offered format matches stated preference | One meeting | Unwanted recording or imposed choice | Meeting end if activated |
**Decision rules:** adopt an invitation only when the person wants it; adapt channel/wording after feedback; repeat only by choice; defer a technique lacking evidence for its intended use. Reject the implementation if it pressures disclosure or ignores a correction, and revise PN-2. Reject a persistent context label and revise PN-3. A positive experience doesn't validate dopamine, a universal decoder or an income claim.
**Drift plan, inactive:** check relevant primary research before teaching decoding as evidence; check a bonus URL only when you need it; re-evaluate a platform-specific claim when the platform/data changes. If these notes are used again after 2027-10-09, reassess the evaluative context before presenting current research status. The EPUB's words, sample counts and 2017 publication facts remain fixed; they don't need recurring reinterpretation as live measurements. No reminder or automation is created.
# Prompt 13 — Visual thinking pack
The maps below are specifications and Mermaid source inside this note. Text equivalents follow each map; color isn't the sole code. Vault roles: green thesis, blue concepts, orange takeaways and purple connections. Full palette retained for future rendering: Russian violet#241c4a; periwinkle#bcb5e3; glaucous#647baf; sky blue#8ecae6; blue-green#219ebc; cerulean#126782; Prussian blue#023047; selective yellow#ffb703; UT orange#fb8500; light orange#fee2c3; light gray#ebebeb. Green#cfe8cf supplies the vault's thesis role. No generated people or inferred demographic identities appear.
## Book and coverage map
Purpose: show three native parts and this review's coverage. Left-to-right layout; blue part nodes, green learning project, dashed sampled node. Sources S-01–16; claims C-001–018. Solid arrows show organization, not proven causal relationships.
```mermaid
flowchart LR
T[Learn interpersonal practices] --> A[First five minutes]
T --> B[First five hours]
T --> C[First five days]
A --> F[Full chapter 3]
B --> G[Full chapters 6 and 9]
C --> H[Full chapter 13]
A -.-> U[Other chapter openings and reviews sampled]
B -.-> U
C -.-> U
classDef thesis fill:#cfe8cf,color:#023047,stroke:#126782;
classDef concept fill:#8ecae6,color:#023047,stroke:#126782;
classDef gap fill:#ebebeb,color:#241c4a,stroke:#647baf,stroke-dasharray:5 5;
class T thesis;
class A,B,C,F,G,H concept;
class U gap;
```
Text equivalent: all three parts contribute to the learning project; chapters 3,6,9,13 receive full text reading. Other chapter evidence is sampled, and unexamined bodies are not evaluated.
## Evidence and dependency map
Purpose: make the weak bridge visible. Top-down layout; observation separate from inference, dashed bridge to contested truth. Sources S-07/S-09–12; C-006/C-009–011; E3/E4. No calibrated effect strength follows from node position.
```mermaid
flowchart TD
O[Observed response or movement] --> Q[Ask what it means in context]
Q --> R[Accept correction and stated preference]
O -.-> X[Known dopamine or hidden truth]
X -.-> W[Universal prediction not established]
P[Native punctuators and photo exceptions] --> Q
classDef practice fill:#fee2c3,color:#023047,stroke:#fb8500;
classDef doubt fill:#ebebeb,color:#241c4a,stroke:#647baf,stroke-dasharray:5 5;
classDef connection fill:#bcb5e3,color:#241c4a,stroke:#647baf;
class O,Q,R practice;
class X,W doubt;
class P connection;
```
Text equivalent: a cue can prompt clarification. The source's possible alternative explanations support context. The dashed hidden-state/prediction path remains uncertain. Asking doesn't guarantee disclosure or agreement.
## Action decision tree
Purpose: keep a person's choice in the process. Top-down, orange practice/decision nodes; claims C-019/C-020 and PN-2/PN-3, source S-15/S-16. This is a system adaptation, not Van Edwards's experiment.
```mermaid
flowchart TD
A[Offer an optional conversation] --> B{Person wants to continue?}
B -->|Yes| C[Ask preferred topic and format]
B -->|No| D[Stop or change topic]
C --> E[Listen and check understanding]
E --> F{Correction or discomfort?}
F -->|Yes| G[Revise or stop]
F -->|No| H[Ask whether help is wanted]
classDef practice fill:#fee2c3,color:#023047,stroke:#fb8500;
classDef decision fill:#ffb703,color:#023047,stroke:#126782;
class A,C,D,E,G,H practice;
class B,F decision;
```
Text equivalent: an invitation allows refusal, continued participation allows correction, and help requires another check. No label or guessed emotional state overrides those responses.
**Omitted visuals:** a causal loop would imply measured neuroscience; a numerical trait dashboard would confer precision the source hasn't established; a historical timeline adds little to the three-stage organization; a separate stakeholder/system map repeats Prompt 8; terminology and comparison matrices already appear in Prompts 3/9. No separate Excalidraw file follows the one-note constraint. Mermaid syntax/fences receive a structural check; actual renderer and Obsidian Sync remain unverified.
**Idea Compass, system navigation:** Parents: popular social-skills teaching and the cited emotion/resource traditions. Children: PN-1/PN-2/PN-3 and optional practice. Friends: responsiveness, accessibility and shared planning. Competitors: hidden-state certainty, single-motive prediction and outcomes judged only by observer impression. These are concepts, not fabricated related-book links.
## Session completion check
Executed 0–13 in order; all produce usable outputs. None intentionally skipped, remaining [], next_prompt null. Adaptations: EPUB as Layer 1; selective analytical reading; qualitative confidence; one canonical note with embedded notes and visuals. No personal rating, reading dates, beliefs or outcomes invented.
Missing material ranked by consequence: (1) independent validation and raw data for universal decoding and transfer; (2) full bodies of the 10 sampled chapters, whose unevaluated evidence could alter a whole-book assessment; (3) reader highlight emphasis; (4) unviewed images, appendix and digital bonuses and original studies not fully read. Research checks state their actual scope, not a full literature audit.
System assumptions are the reading contract, clusters, repairs, ethical standards, permanent-note candidates, invented case and experiments. Source classes retain EA/Q3P/RI. The authoritative 2017 publication date differs from the OPF package timestamp; no prior fact is silently replaced. Canonical Markdown persistence can be checked by hash. Live Obsidian rendering and Sync are unverified; shared log/queue verification and temporary cache cleanup remain the root agent's responsibility.